Patients with acute hypoxemic respiratory failure (ARDS) associated with severe noncardiogenic pulmonary edema exhibit a depressed ability to extract oxygen in the periphery. In these patients, an increase in O2 delivery (QO2 = cardiac output x arterial O2 content) caused by a change in cardiac output (Qt) is associated with an increase in O2 uptake (VO2). Published studies of patients with ARDS suggest that systemic O2 delivery of 21 ml/min/kg is necessary to maintain VO2 independent of QO2. By contrast, patients without respiratory failure show no dependence of VO2 on QO2 unless delivery fails below 8.2 ml/min/kg. Hence, patients with ARDS exhibit supply dependence of VO2 at an extraction ratio ((CaO2-CVO2)/CaO2) of .3 - .4,; patients without respiratory failure can achieve extraction rations of .6 - .7 without introducing QO2 dependence. An inability of peripheral tissues to extract sufficient O2 to maintain aerobic metabolism may be caused by a maldistribution of VO2/QO2 at a regional tissue level, or within tissue beds. Alternatively, increased diffusion distance between capillary and mitochondria may limit O2 unloading by diffusion. It is also possible that a pathologic supply dependence of VO2 on QO2 may occur if hypoxic tissue damage causes O2 uptake to increase via superoxide radical formation. Experimental studies are proposed to test regional tissue O2 extraction in anesthetized canine isolated hind limb, isolated small bowel segment, and isolated kidney preparations relative to whole body. This approach will clarify relative contributions of interregional vs intraregional blood flow maldistribution as mechanisms contributing to a pathologic supply dependence. I propose to compare the extraction limit in normal animals with that in two experimental models demonstrating whole body abnormal dependence of VO2 on QO2. Specifically, mechanisms responsible for a disordered extraction capacity during Pseudomonas aeruginosa bacterial infusions or ventilation with high inspired O2 fractions will be explored. Additionally, I propose to explore the value of multiple indicator dilution assessment of permeability surface area products in our isolated tissue models to clarify the contribution of changes in the recruitable capillary surface area as a mechanism for altered extraction capacity. These studies will help to clarify the pathophysiological mechanisms responsible for a dependence of VO2 on QO2, and will help to uncover potentially therapeutic interventions for the maintenance of peripheral O2 extraction in patients with acute hypoxemic respiratory failure.